Altering a computerized laboratory test order form rationalizes ordering of laboratory tests in primary care physicians

• Reorganizing a computerized laboratory test ordering form reduced use of the tests.
• This led to GPs more thoroughly considering the use of laboratory tests.
• This intervention reduced ordering of inappropriate tests by up to 80–90%.
• It modulated the clinical behavior of GPs in a highly sustainable manner.

BackgroundTo reduce physicians’ inappropriate laboratory requests for their patients, administrators have used methods such as modifying a laboratory request order form with an agreed requesting protocol for the most common diagnoses in primary health care.ObjectiveTo study the effects of removing the erythrocyte sedimentation rate (ESR) and aspartate transaminase (AST) which are considered of limited clinical value for primary care clinical decision-making from a computerized laboratory test order form. These tests were removed to another new view from the electronic laboratory menu where the physicians, instead of just ticking the desired test from the list, had to do 4–8 s extra work by writing down the abbreviation to order the test.MethodsAn observational controlled prospective study based on a before-after design was performed by removing AST and ES from the laboratory test order form of the computerized laboratory system for all primary care in the city of Helsinki, Finland. The numbers of annual and monthly use of AST and ESR and their controls, alanine transaminase (ALT) and C-reactive protein (CRP) ordered by General practitioners (GPs) was recorded over an eight-year period: four years before and a four years after the removal of AST and ES.ResultsRemoving AST and ESR from the computerized laboratory test order form decreased their use by up to 90%, whereas the use of the control tests increased throughout the follow-up period. The variation in use of these removed tests also decreased.ConclusionRemoving a laboratory test from a computerized laboratory test order form may significantly reduce GPs’ use of the laboratory test. Further studies are needed, however, to ensure the safety of this type of intervention.